Statistical Detection and Characterization of a Deviation from the Gutenberg-Richter Distribution above Magnitude 8

— We present a quantitative statistical test for the presence of a crossover c₀ in the Gutenberg-Richter distribution of earthquake seismic moments, separating the usual power-law regime for seismic moments less than c₀ from another faster decaying regime beyond c₀. Our method is based on the transformation of the ordered sample of seismic moments into a series with uniform distribution under condition of no crossover. A simulation method allows us to estimate the statistical significance of the null hypothesis H₀ of an absence of crossover (c₀=infinity). When H₀ is rejected, we estimate the crossover c₀ using two different competing models for the second regime beyond c₀ and the simulation method. For the catalog obtained by aggregating 14 subduction zones of the Circum-Pacific Seismic Belt, our estimate of the crossover point is log(c₀)=28.14 ± 0.40 (c₀ in dyne-cm), corresponding to a crossover magnitude m_W=8.1 ± 0.3. For separate subduction zones, the corresponding estimates are substantially more uncertain, so that the null hypothesis of an identical crossover for all subduction zones cannot be rejected. Such a large value of the crossover magnitude makes it difficult to associate it directly with a seismogenic thickness as proposed by many different authors. Our measure of c₀ may substantiate the concept that the localization of strong shear deformation could propagate significantly in the lower crust and upper mantle, thus increasing the effective size beyond which one should expect a change of regime.